 Daniel Bevenius
|
a6e514a85f
llava: fix typo/formatting in README.md (#5405)
|
1 anno fa |
 Johannes Gäßler
|
26d4efd11e
sampling: fix top_k <= 0 (#5388)
|
1 anno fa |
 Georgi Gerganov
|
8504d2d0da
tests : .gitignore obj files
|
1 anno fa |
 Michael Podvitskiy
|
c4fbb6717c
CMAKE_OSX_ARCHITECTURES for MacOS cross compilation (#5393)
|
1 anno fa |
 Ebey Abraham
|
8c933b70c2
fix typo in readme (#5399)
|
1 anno fa |
 Kamil Tomšík
|
b906596bb7
Add Ava in the list of llama.cpp UIs (#4362)
|
1 anno fa |
|
Johannes Gäßler
|
aa7ab99be2
CUDA: fixed mmvq kernel for bs 2,3,4 and -sm row (#5386)
|
1 anno fa |
 Neo Zhang Jianyu
|
10afa6f1d1
[SYCL] update install make by w64devkit (#5297)
|
1 anno fa |
 Xiao-Yong Jin
|
0ef46da632
llava-cli : always tokenize special tokens (#5382)
|
1 anno fa |
 0cc4m
|
ee1628bdfe
Basic Vulkan Multi-GPU implementation (#5321)
|
1 anno fa |
 Eve
|
ed0bf32290
readme : modernize (#5379)
|
1 anno fa |
 Ben Williams
|
9a697d842b
readme : update ui list (#5354)
|
1 anno fa |
 runfuture
|
316c7faf77
llama : add MiniCPM support (#5346)
|
1 anno fa |
 Justin Parker
|
f3e2b4fa3f
server : update `/props` with "total_slots" value (#5373)
|
1 anno fa |
 Sang-Kil Park
|
f68664ac24
convert : fix TypeError on GPT-2 vocab.json (#5288)
|
1 anno fa |
 Alexey Parfenov
|
213d1439fa
server : remove model.json endpoint (#5371)
|
1 anno fa |
|
Johannes Gäßler
|
17c97fb062
CUDA: mul_mat_vec_q max. batch size 8 -> 4 (#5370)
|
1 anno fa |
 Kawrakow
|
b08f22c882
Update README.md (#5366)
|
1 anno fa |
|
Kawrakow
|
f57fadc009
Slight quantization improvement for Q4_K and Q5_K (#5361)
|
1 anno fa |
 BarfingLemurs
|
2e9c0bd6b3
readme : add phi, orion 14b, internlm2, and yi-VL to readme (#5362)
|
1 anno fa |
|
Johannes Gäßler
|
2c516611f1
CUDA: mul_mat_vec_q for batch sizes > 1 (#5351)
|
1 anno fa |
|
Justin Parker
|
8a79c591de
server : include total "num_slots" in props endpoint (#5349)
|
1 anno fa |
 Michael Coppola
|
31e7903221
server : add `dynatemp_range` and `dynatemp_exponent` (#5352)
|
1 anno fa |
 Niall Coates
|
4ffc7a17d4
server : various fixes for the prompt field in /completion (#5300)
|
1 anno fa |
|
Georgi Gerganov
|
906cff55c2
py : handle byte tokens in `get_token_type` (#5341)
|
1 anno fa |
|
Johannes Gäßler
|
098f6d737b
make: Use ccache for faster compilation (#5318)
|
1 anno fa |
|
Johannes Gäßler
|
78b00dda6c
README: updated introduction (#5343)
|
1 anno fa |
|
Kawrakow
|
c6b395535a
ggml : make use of ggml-quants.h possible in C++ code (#5338)
|
1 anno fa |
 Dr. Tom Murphy VII Ph.D
|
abb61944a5
ggml : avoid duplicating function calls using MIN/MAX macros (#5325)
|
1 anno fa |
|
Kawrakow
|
89503dcb5f
iq3_xxs: quards for the no-imatrix situation (#5334)
|
1 anno fa |
